Richard Paige – Maybe it was the Nobel Prize, or my wanting to become a bit more familiar with things here on campus in what is week three for me at Wabash. Perhaps it was the lure of free pizza.
No matter the reason, I swallowed hard, and climbed the stairs to Hays 319 for what was to be my first science lecture in a quarter century.
I was curious, and isn’t that half the battle?
West Virginia University’s Dr. Blake Mertz was giving a noontime talk on computational chemistry entitled, “Novel PIP2 Lipid Binding to Focal Adhesion Kinase Probed Through Coarse-Grained Molecular Dynamics Simulations.”
Computational chemistry is a branch of the science that uses the computer to help predict how molecules interact, which can be essential to understanding the nature of diseases and the production of medicine.
Last week, three scientists – Martin Karplus, Michael Levitt and Arieh Warshel – shared the Nobel Prize for chemistry with pioneering work in this area, much of it dating back to the early 1970s.
Feller continued by saying that the average person just wants science to come up with a cure for cancer so to speak, but that Karplus, Levitt and Warshel, “enabled others to find that cure with a new way to think about and study a problem.”
Mertz’s talk opened with some basics, that amino acids are the building blocks of proteins. Check. I remember that from way back when. And much of what was discussed today, i.e. molecular dynamics, would be based on Newton’s second law of motion (F=ma). Check. I remember that, too.
From that point, much of what was discussed went clear over my head. The simple thought is that one still needs to understand the theory behind the problem, how the forces change in an experiment, before all is plugged into the computer and the results are produced.
Before all of this can happen, four questions need to be answered:
1.) Can I answer my question using simulation? The simpler the question, the more likely that this is the case.
2.) What is the timescale of what I want to observe?
3.) How accurate does my model need to be?
4.) Do I have a computer that can handle the calculations?
What I found as interesting was the fact that as these computer simulations become more powerful, there remains the potential that such simulations could render traditional experiments obsolete.
“I think eventually we’ll find a happy medium,” said Mertz. “There is still huge bias toward experimental results in the scientific community, like ‘If I can’t see it, then I’m not going to believe it.’
“Certainly, these are questions that need to be answered,” he continued. “The benefit of doing simulations is that you can really look at things on an atomic level. Experimentalists are trying to get to that level where they can see atoms, but that level is so small that you can’t answer those questions with an experiment. Ultimately, I think that computational chemistry is another tool for us to use. What separates the men from the boys in the lab is how judiciously you use them.”
Now that’s language that I can understand.